client log is attached.

Deleted: client.0.DEBUG 16.8 KB

client.0.DEBUG 16.8 KB Download

I submitted the CL to move grahics_Gbm out of bvt-inline.

The peach_pit and nyan_big failures happened when various vgem related CLs were in the CQ:

https://luci-milo.appspot.com/buildbot/chromeos/nyan_big-paladin/4078

the gbmtest uses vgem, hence the various failures.  These changes are now out of the CQ.  Moving graphics_Gbm out of bvt-inline would not the correct course of action in this case.

Looks like <https://chromium-review.googlesource.com/c/chromiumos/third_party/autotest/+/861691> was marked -1.

Can we make sure that the vgem CLs are marked -1 until this gets resolved?

OK, marking -1 on:

* https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/858263
  BACKPORT: CHROMIUM: drm/vgem: Add short circuit for reimporting our GEMs
  kernel 3.14

* https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/858490
  BACKPORT: CHROMIUM: drm/vgem: Add short circuit for reimporting our GEMs
  kernel 3.10

* https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/858599
  BACKPORT: CHROMIUM: drm/vgem: Add short circuit for reimporting our GEMs
  kernel 3.8

Other vgem changes int the kernels depend on those ones so it should keep the others out.  Even though 3.14 devices didn't fail, still seems best to mark them all -1.  

---

The -1 verified should get cleared by a rebase, or I'm happy to remove the -1 as needed.

===

Note that I _didn't_ mark this as -1 even though it's vgem related:

* https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/860737
  CHROMIUM: add case of CONFIG_ARM64 in drm_clflush_sg
  kernel 3.18

Thank you for locating. I abandoned the CL to move graphics_Gbm out of bvt-inline.

Fixed the issues with the vgem patches.